Monday, March 28, 2011

The Uselessness of Predictions

Whole industries have sprung up over the centuries for the mere purpose of predicting the future, from the vestal virgins of Delphi and witches of Scotland to star gazers, face/palm/foot readers, statistical manipulators, scientific safeguarders, trend discerners and inspired third-sense charlatans.

When all potential faults and problems are properly identified and resolved, everything goes on so smoothly that it is presumed that that which have been hard at work and keeping everything in tip-top condition are but irritations because of their apparent presumed self-importance which is making the incompetents extremely insecure and jealous.

There is really no reward in forecasting except when one is making plans for one's own life journey. Even then, there are those who argue that there is no need for planning and everything will end up in the best of all possible worlds, so long as we accept whatever comes. For those who have no experiences of good things coming by themselves on their own accord, without effort on one's part, this is an uneasy and possibly traumatic attempt at relaxation. In an organisation or society, the politicians will always short-change the technical people in order to exert control over the experts. Under such repressive conditions, it is likely that the technical upkeep may just be a notch below the necessary as a form of sabotage in retaliation to presumed injustice. Such perverse behaviour may even be observed in children when they underperform in order to hurt their parents. First hand, I tend to try to identify potential problems long before they surface, so that easier milder remedial actions can be applied earlier, but have more often than not been labelled as a person who sees problems where none exist (for the moment). My standard response is this: "I hope I am wrong. If I am not, you are in deep trouble."

But there are things in life that are self-evident who do not need a rocket scientist to predict and which every sensible ordinary person with some common sense will tell you will happen, as a matter of fact: (1) You will die. (2) Human beings will disappear from earth, just as dinosaurs had. (3) You make bombs, bombs kill people. (4) Nuclear plants will blow up.

For natural phenomena such as (1) and (2), there is nothing much that one can do except to accept the inevitable, although most of us will still try to stretch things at the margin just a little more (just like home renovation). You find people popping pills (me included), exercise, and resorting to all kinds of activities of some sort, wasting time in the hope of gaining some, which is nothing but the swapping of young energetic time for a longer weak and frail condition.

Man-made phenomena such as (3) and (4) are simply the foolishness of human beings, of a small group of people harming the whole of mankind just because society at large allow a few foolish people to take control over the majority just because the few claims to have the right or sheer brute force to exert that power.

Will there ever be wisdom at the top, when the foolish thinks that status and position bestow upon them wisdom which they do not possess. The jostling for power and the scrambling for titles are things that foolish people do to feel self important. Even erstwhile perfectly good individuals when they are feeling ordinary and are wise seem to have lost their wits the moment they enter halls of "monumental historical importance." When they do not deserve the position, we will find that they generally will destroy the sacredness of the place through the corruption of its integrity.

Because of the foolishness of individuals, we can only allow each individual a short period of time of time to display their foolish the moment they step into power.

4 comments:

semuanya OK kot said...

Modelling the future began with social inteaction, migration and religion.

- Our addiction to pollsters and forecasters is a symptom of our chronic uncertainty. Even when the forecasts prove wrong, we go on watching them read the entrails of statistical tables and graphs. - Eric Hoffer, adapted

- The difference between a cab driver and a history professor is only cosmetic; the latter expresses himself better. The social sciences – economics, sociology, political science – and humanities provide little objective understanding of history. Newspapers, history books, analyses and economic reports provide only misplaced confidence regarding causes. There is partial empirical evidence of this in the disciplines that offer quantitative data and predictions. Economists fare hardly better than random in forecasting inflation, growth, interest rates, balance of payment, etc. Yet we depend on them for governmental economic policy! Those who argue that the forecasts might impact these variables should consider the mechanism of "self-canceling prophecy". In the 1960s, Victor Niederhoffer conducted impressive relevant tests on how the news affects prices. The experts tend to predict only the regular events – though quite poorly at that – while missing the large "unusual" events. While economists get caught soon, news analysts, biographers, and other experts get away for longer. - Prof. Nassim Taleb, edge.org, 2005, adapted

- Modern science is typically based on a divide based on a set confidence level, stripped of consequences. Probability by itself is a non-observable academic construction. Using it in a standalone way has caused harm in many fields. It is expectation - the product of probability and payoff - or a series of expectation values that matters. On occasion, the potential payoff can be so vast that it dwarfs the probability - which may itself not be computable. Besides simple games of chance, you almost never face a single probability with a single and known payoff. You cannot rely on statistical knowledge for non-standard decisions. For the risk of an earthquake, you may face a 5% probability of magnitude 3 or higher, a 2% probability of magnitude 4 or higher, etc. Seeking the "probability of an earthquake" is a meaningless question for risk assessment. This is the misconception that helps the banking system go bust with an astonishing regularity. I have shown that institutions exposed to negative rare events, like banks and some classes of insurance ventures, have almost never been profitable over long periods. In the current financial mess, not only were the quants and other pseudo-experts in bank risk management wrong about the probabilities, they were severely wrong about the different layers of depth of potential negative outcomes. For instance, Morgan Stanley has lost about $10 billion so far despite having foreseen the crisis and executed hedges against it; they just did not realize how deep it would be. In the current debate on carbon emissions and climate change, even if the claims of impending danger were shoddy science, you should consider the consequences of the very remote possibility that they may be right. - Prof. Nassim Taleb, edge.org, 2008, adapted

1 of 2

walla said...

Sometimes weekends are bad so that monday mornings bring an unpredictable storm or two.

Life isn't about how to survive such storms but how to dance in their rain.

And the happiest people don't necessarily have the best of everything all the time. Just that they make the best of everything they have at any time.

Such as still being able to kick up a ruckus when one sees something is not right.

And generally things can get bad or stale because we can't predict the future well enough to remove the anxiety of the present.

Just about the only thing we can predict is the past and even that we can't be too sure whether the data collection, processing and analysis haven't been tardy, incomplete or imprecise.

Of course, it is useful to be able to predict how things will pan out within a certain range by a certain date and at a certain confidence level, all the more so when heaps of money are at stake.

In such a situation, the punter will only play by ear and make a quick decision depending whether at the point of making the decision he thinks it was news and not rumor, and how much was its respective weightage on the plausible direction of the event, ceterus meerebus aside.

On the other hand, the professional would have weighed in with econometric models and time series, what more hunching wisdom from years of observation.

If the punter is right, it was because he was lucky. If the professional is right, it was because he had executed a defensible and repeatable method.

If the punter is wrong, he would be wrong for the same reason he was right and that's the flip of the coin. If the professional is wrong, it was because the circumstances under which his method would otherwise have been relevant have changed.

Take the lunar-buff who had preannounced the Christchurch quake. He bravely assayed another prediction so that people took to the streets in advance only to find the day came and went without fanfare leaving them to pummel his name for trying to predict something which had only resulted in their inconvenience. But what if he had been right a second time?

Having dispensed with the homilies, one must admit that uselessness does have its usefulness. For instance, one can be useless to a bad boss or organization and that can be considered useful to society as a result.

Sometimes it's the notion which is bad so that if one is useless to a bad notion, the uselessness is actually useful in another regard to the extent that less damage is done.

In other words, if the mug is already half full, why not fill it to the brim?

In the past weeks, and maybe taking a leaf from the lunar-buff that it's all about lunar effects, you can't escape the suspicion that some of the notions surfacing have been predictably bad. Maybe that's why the decepticons are hiding on the dark side of the moon.

Good can return either good or bad. Bad can return either bad or good. No prediction machine needed there.

But there is a small, teeny, weeny, minute, asymmetry in between both.

Blessed are those who can catch it. Although a will-o'-wisp, it's actually a small beacon signal from somewhere else, a predictor in its own right.

Maybe that's the only reason why man has failed in the fine art of making predictions with an unfailing constancy of success.

walla said...

As an exercise in prediction, though not of the economic kind, consider this:

1. the authorities have removed themselves from pre-recognizing medical qualifications because they said they cannot afford to go around and accreditate the private institutions.

2. this means they will only issue practicing licenses if and when the graduates pass certain exams after graduating from those private institutions.

3. this opens up choices but it also opens up opportunities for poorly qualified institutions to make big money out of desperate families;

4. anyone going to an education fair will see how commercially competitive this sector is; temporary staff are paid to recruit students, often on commission basis so that commercial interest can supersede the delivery of accurate information for better decision-making, or the watering down of adverse information.

There is already one case highlighted of a substantial nonrefundable fee paid only for the student to find the medical classes are too few in a day which means the programme is too basic for such a course.

Now imagine parent and child rushing to meet deadlines to enroll, coaxed to make hefty nonrefundable payments in advance, signing on very fine print, and then the youngster goes through six years of a course in an institution with subpar facilities including lack of properly trained and qualified lecturers so that he or she comes out years later and can't understand why it is so difficult to pass the licensing exams.

So, unlike lawyers who can find income doing other things, what does such a privately trained medical graduate do until such time, if ever, the licensing exam is passed? All this apart from the risk of wasting half a million ringgit and six years of study?

This particular profession calls for certain irrevocable standards across board and the authorities must not expect the public to know anything about the profession - from the quality of lecturers qualifications and the type of facilities and level of coursework - to a depth essential to make a hard decision whether the youngster is to pursue the course and make an investment to a private institution which is not recognized or accreditated in advance.

The authorities must realize this yesterday and issue a serious public education program on what they are supposed to look out for

- how many hours lecture in a week for what subjects;
- what sort of initials lecturers should carry;
- what are the practicals needed;
- what are the names of the textbooks to be used;

apart from the fees schedule and examinations requirements etc.

and then make those recruiting institutions responsible to issue such information before the parent or student signs up.

That this is not done when the policy was changed is recipe for disaster.

I hope someone is reading this and can say whether this prediction has merit.

pooja nabar said...

i personally feel predicting future by intuition in the case of human beings & by a sophesticated technique like delphy in an buziness organisation, this gives a purpose to live for that person, & make future plans, to deal with the predicted future consequences, for the organisation.